Subscribe to EDOM TECH Newsletter
2026/03/25
Building Intelligent Machines: NVIDIA® Jetson Thor™ at the Core of Physical AI
Building Intelligent Machines: NVIDIA® Jetson Thor™ at the Core of Physical AI
From Research to Real-World Intelligence
Physical AI is rapidly moving from research labs into real-world deployment. From autonomous robots in warehouses to intelligent machines operating on factory floors, a new generation of systems is emerging that can perceive, reason, and act in the physical world. At the heart of these systems lies the compute platform that transforms sensor data and AI models into real-time decisions. This is where NVIDIA® Jetson Thor™ comes in as the next-generation edge AI platform designed to power physical AI and robotics. Jetson Thor™ brings together high-performance GPU architecture, accelerated AI pipelines, and a robust software ecosystem to support large multimodal models, real-time perception, and advanced decision-making workloads. For robotics and autonomous systems, this enables machines to interpret sensor inputs, understand their surroundings, and act with higher levels of autonomy, transforming perception and planning models into real-world intelligence.
Powering Systems with a Complete Edge Ecosystem
To fully realize physical AI systems, the compute platform must be supported by a robust ecosystem of peripheral technologies. High-performance storage enables rapid access to large AI models and continuous sensor data streams, while advanced connectivity integrates edge systems with cloud infrastructure. High-speed I/O interfaces allow seamless connection to cameras, sensors, and industrial equipment used in robotics and autonomous machines. Sensor bridge technologies further enhance these systems by aggregating and synchronizing inputs from cameras, LiDAR, radar, and other sensors, delivering structured data into the compute pipeline. Vision and perception modules accelerate tasks such as object detection, spatial understanding, and environment mapping, enabling machines to interpret and navigate complex real-world environments.
Enabling Scalable Physical AI at the Edge
Together, these technologies—from storage and connectivity to high-speed I/O, sensor bridging, and perception—form a complete physical AI ecosystem. At its center, NVIDIA® Jetson Thor™ provides the compute engine that brings sensor data, AI models, and real-time decision-making together. By delivering powerful AI computing at the edge, Jetson Thor™ enables intelligent machines to move from simulation into real-world operation with greater autonomy, efficiency, and scale.
Featured Products
Storage & Wireless Connectivity
- Silicon Motion SM8008 Series - PCIe Gen5 SSD Controller
- Kingston M.2 2280 - TLC PCIe NVMe SSD
- Ampak AC5035M2 - 5G RedCap Cellular Module
- Telit FN990 - LTE/5G Data Card
- Ampak AP12275_M2P - Wi-Fi and Bluetooth 2230 M.2 Card
- Telit WE310K6-P - Dual-Band Wi-Fi and BLE5 Combination Module
High-Speed I/O & Sensor Bridge
- Marvell AQC113C - Scalable mGig Ethernet Controller
- Marvell AQR113C - Single-port Ethernet PHY Transceiver
- Microchip MPF200T/100T - PolarFire FPGA
Camera Vision & Perception Module
- Framos FSM:GO IMX Series MIPI Camera Module
- Framos D400e Series - Industrial Depth Camera
- oToBrite oToCAM223-H61M - 2.95M Automotive Camera
- oToBrite oToCAM271ISP-C120M - 8MP Automotive Camera
- Intel RealSense D435i - Stereo Depth Camera
